• ETRI Journal
• /
• v.46 no.2
• /
• pp.263-276
• /
• 2024

Different diseases occur in the brain. For instance, hereditary and progressive diseases affect and degenerate the white matter. Although addressing, diagnosing, and treating complex abnormalities in the brain is challenging, different strategies have been presented with significant advances in medical research. With state-of-art developments in artificial intelligence, new techniques are being applied to brain magnetic resonance images. Deep learning has been recently used for the segmentation and classification of brain images. In this study, we classified normal and pathological brain images using pretrained deep models through transfer learning. The EfficientNet-B5 model reached the highest accuracy of 98.39% on real data, 91.96% on augmented data, and 100% on pathological data. To verify the reliability of the model, fivefold cross-validation and a two-tier cross-test were applied. The results suggest that the proposed method performs reasonably on the classification of brain magnetic resonance images.

• Journal of radiological science and technology
• /
• v.35 no.2
• /
• pp.133-139
• /
• 2012

For the examination of nuclear medicine, myocardial scan is a good method to evaluate a hemodynamic importance of coronary heart disease. but, the automatized qualitative measurement is additionally necessary to improve the decoding efficiency. we suggests the creation of cardiac three-dimensional model and model of three-dimensional cardiac thickness as a new measurement. For the experiment, cardiac reduced cross section was obtained from SPECT. Next, the pre-process was performed and image segmentation was fulfilled by level set. for the modeling of left cardiac thickness, it was realized by applying difference equation of two-dimensional laplace equation. As the result of experiment, it was successful to measure internal wall and external wall and three-dimensional modeling was realized by coordinate. and, with laplace formula, it was successful to develop the thickness of cardiac wall. through the three-dimensional model, defects were observed easily and position of lesion was grasped rapidly by the revolution of model. The model which was developed as the support index of decoding will provide decoding information to doctor additionally and reduce the rate of false diagnosis as well as play a great role for diagnosing IHD early.

• Journal of Korean Society of Transportation
• /
• v.28 no.2
• /
• pp.151-159
• /
• 2010

This paper presents a research result that was performed to develop a more accurate freeway crash prediction model than existing models. While the existing crash models only focus on developing crash relationships associated with highway geometric conditions found on a short section of a crash site, this research applies a different approach considering the upstream highway geometric conditions as well. Theoretically, crashes occur while motorists are in motion, and particularly at freeways vehicle speed at one specific point is very sensitive to upstream geometric conditions. Therefore, this is a reasonable approach. To form the analysis data base, this research gathers the geometric conditions of the West Seaside Freeway 269.3 km and six years crash data ranging 2003-2008 for these freeway sections. As a result, it is found that crashes fit well into Negative Binomial Distribution, and, based on the developed model, total number of crashes is inversely proportional to highway curve length and radius. Contrarily, crash occurrences are proportional to tangent length. This result is different from existing crash study results, and it seems to be resulted from this research assumption that a crash is influenced greatly by upstream geometric conditions. Also, this research provides the expected effects on crash occurrences of the length of downgrade sections, speed camera placements, and the on- and off- ramp presences. It is expected that this research result is useful for doing more reasonable highway designs and safety audit analysis, and applying the same research approach to national roads and other major roads in urban areas is recommended.

• Journal of KIISE:Software and Applications
• /
• v.33 no.1
• /
• pp.95-104
• /
• 2006

The Gradient Vector Flow (GVF) snake or active contour model offers the best performance for image segmentation. However, there are problems in classical snake models such as the limited capture range and the slow progress into concavity. This paper presents a new method for enhancing the performance of the GVF snake model by extending the external force fields from the neighboring fields and using a modified smoothing method to regularize them. The results on a simulated U-shaped image showed that the proposed method has larger capture range and makes it possible for the contour to progress into concavity more quickly compared with the conventional GVF snake model.

• Journal of Korean Society on Water Environment
• /
• v.29 no.2
• /
• pp.212-222
• /
• 2013

A watershed model was constructed using Hydrological Simulation Program Fortran to quantitatively predict the stream flows at major tributaries of Nakdong River basin, Korea. The entire basin was divided into 32 segments to effectively account for spatial variations in meteorological data and land segment parameter values of each tributary. The model was calibrated at ten tributaries including main stream of the river for a three-year period (2008 to 2010). The deviation values (Dv) of runoff volumes for operational stream flow forecasting for a six month period (2012.1.2 to 2012.6.29) at the ten tributaries ranged from -38.1 to 23.6%, which is on average 7.8% higher than those of runoff volumes for model calibration (-12.5 to 8.2%). The increased prediction errors were mainly from the uncertainties of numerical weather prediction modeling; nevertheless the stream flow forecasting results presented in this study were in a good agreement with the measured data.

• Proceedings of the KOSOMBE Conference
• /
• v.1997 no.05
• /
• pp.116-118
• /
• 1997

In hippocampal morphology Abnormalities, including unilateral or bilateral volume loss, are known to occur in epilepsy, Alzheimer's disease, and in certain amnestic syndromes. To detect such abnormalities in hippocampal morphology, we present a method that combines region growing and dynamic contour model to detect hippocampus from MRI brain data. The segmentation process is performed two steps. First region growing with a seed point is performed in the region of hippocampus and the initial contour of dynamic contour model is obtained. Second, the initial contour is modified on the basis of criteria that integrate energy with contour smoothness and the image gradient along the contour. As a result, this method improves fairly sensitivity to the choice of the initial seed point, which is often seen by conventional contour model. The power and practicality of this method have been tested on two brain datasets. Thus, we have developed an effective algorithm to extract hippocampus from MRI brain data.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.12
• /
• pp.1216-1222
• /
• 2009

This article describes a simple method for generating the walking trajectory for the biped humanoid robot. The method used a simple inverted model instead of complex multi-mass model and a reasonable explanation for the model simplification is included. The problem of gait trajectory generation is to find the solution from the desired ZMP trajectory to CoG trajectory. This article presents the analytic solution for the bipedal gait generation on the bases of ZMP trajectory. The presented ZMP trajectory has Fourier series form, which has finite or infinite summation of sine and cosine functions, and ZMP trajectory can be designed by calculating the coefficients. From the designed ZMP trajectory, this article focuses on how to find the CoG trajectory with analytical way from the simplified inverted pendulum model. Time segmentation based approach is adopted for generating the trajectories. The coefficients of the function should be designed to be continuous between the segments, and the solution is found by calculating the coefficients with this connectivity conditions. This article also has the proof and the condition of solution existence.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2003.06a
• /
• pp.1634-1637
• /
• 2003

Any given object on the motor-driven turntable is pictured from 8 to 72 different views with a digital camera. 3D shape reconstruction is performed with the integrated software called by Scanware from these multiple digital photographs. There are several steps such as configuration, calibration, capturing, segmentation, shape creation, texturing and merging process during the shape reconstruction process. 3D geometry data can be exported to cad data such as Autocad input file. Also 3D image model is generated from 3D geometry and texture data, and is used to advertise the model in the internet environment. Consumers can see the object realistically from wanted views by rotating or zooming in the internet browsers with Scanbull spx plug-in. The spx format allows a compact saving of 3D objects to handle or download. There are many types of scan equipments such as laser scanners and photogrammetric scanners. Line or point scan methods by laser can generate precise 3D geometry but cannot obtain color textures in general. Reversely, 3D image modeling with photogrammetry can generate not only geometries but also textures from associated polygons. We got various 3D image models and introduced the process of getting 3D image model of an internet-connected watchdog robot.

• ETRI Journal
• /
• v.33 no.3
• /
• pp.423-433
• /
• 2011

This paper introduces an adaptive and integrated utterance verification (UV) framework using minimum verification error (MVE) training as a new set of solutions suitable for real applications. UV is traditionally considered an add-on procedure to automatic speech recognition (ASR) and thus treated separately from the ASR system model design. This traditional two-stage approach often fails to cope with a wide range of variations, such as a new speaker or a new environment which is not matched with the original speaker population or the original acoustic environment that the ASR system is trained on. In this paper, we propose an integrated solution to enhance the overall UV system performance in such real applications. The integration is accomplished by adapting and merging the target model for UV with the acoustic model for ASR based on the common MVE principle at each iteration in the recognition stage. The proposed iterative procedure for UV model adaptation also involves revision of the data segmentation and the decoded hypotheses. Under this new framework, remarkable enhancement in not only recognition performance, but also verification performance has been obtained.

• Convergence Security Journal
• /
• v.21 no.3
• /
• pp.49-56
• /
• 2021

Recently, artificial intelligence is regarded as an essential technology in our society. In particular, the invasion of privacy in artificial intelligence has become a serious problem in modern society. Split learning, proposed at MIT in 2019 for privacy protection, is a type of federated learning technique that does not share any raw data. In this study, we studied a safe and accurate segmentation learning model using known differential privacy to safely manage data. In addition, we trained SVHN and GTSRB on a split learning model to which 15 different types of differential privacy are applied, and checked whether the learning is stable. By conducting a learning data extraction attack, a differential privacy budget that prevents attacks is quantitatively derived through MSE.